Predictive modelling of Salmonella spp. inactivation in pork burger patties of varying fat contents

Salmonellosis is the second most common reported cause of foodborne illness annually in Australia with pork products implicated in outbreaks (Pointon in press). Pork burgers are a serving option for pork mince and can potentially have foodborne pathogens internalised during grinding. Predictive models exist for the thermal inactivation of E. coli O157:H7 in beef patties (Juneja et al. 1997), but analogous models for Salmonella spp. in pork patties could not be found. This study aimed to create such a model and determine if fat content or serovar affect Salmonella survival.

Pre-packaged pork minces marked as ‘Regular’ and ‘Extra Lean’ were purchased from a retail chain in Adelaide, inoculated with one of three Salmonella serovars (S. 4,[5],12,i:–, S. Senftenberg and S. Typhimurium; all isolated previously from porcine sources), and formed into pork burger patties. Before inoculation, samples of the mince were taken for fat content determination by fatty acid extraction. Patties were then cooked to various internal endpoint temperatures on an electric skillet, bagged and rested for 3 min before being submerged in ice. Patties were then homogenised in buffered peptone water, with serial dilutions of the homogenate plated on Xylose Lysine Deoxycholate agar plates, incubated at 37°C for 22 ± 2 h and typical colonies counted. In total, 144 patties were formed, 126 were cooked and 18 uncooked controls over 18 experiments. Data on the internal endpoint temperature (°C), fat content of the mince and Salmonella serovar, were fitted to the three parameter logistic regression model of Wadley (1949) scaled to the concentration in the raw patties to generate a predictive model for Salmonella concentration (CFU/g). The overall mean fat content was 6.11%, but two distinct groups were observed: <5% fat (mean 2.99%) and >10% fat (mean 12.35%), i.e. the fat content of mince samples did not correspond to the nomenclature used on the packages; some batches had lower fat contents than indicated. Separate models were developed for each of these groups and, within each group, fat content was treated as a continuous variable. Interactions between the temperature and fat content influenced Salmonella survival (P = 0.043). The difference between fat groups disappeared as the temperature approached 62°C (Fig. 1). For pork mince with mean fat contents of 2.99% and 12.35%, Salmonella survival was predicted to decrease by 0.227 and 0.268 log₁₀ CFU/g respectively for a 1°C increase in temperature. For both fat groups, a 5-log₁₀ reduction in the Salmonella concentration was predicted to occur at 63°C. There were no significant differences in the inactivation kinetics between the three serovars (P > 0.05).

Fig. 1.  Salmonella spp. concentrations at each endpoint cooking temperature. The predictive model for the means of the two groups of mince (<5% fat, solid line and >10% fat, dashed line) is also depicted.

S. 4,[5],12,i:– is an emerging serovar of public health interest (Pointon in press) and as revealed in this study, it appears to have a similar inactivation kinetics to other serovars. A novel predictive model was developed for inactivation of Salmonella spp. in pork burgers, using the model by Wadley (1949), not previously used in this context. Reduced fat in pork burger patties may decrease the risk of salmonellosis when cooked to a lower degree of ‘doneness’. This work provides industry with knowledge that can be used in marketing to inform consumers about pork burger cooking and in food service to validate safe cooking processes.